Run 3. Outer Step 6. Inner Step 260.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5389308,
-    5389311,
-    5389315,
-    5389319,
-    5389322
+    5389354,
+    5389358,
+    5389361,
+    5389365,
+    5389368
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 258,
+  "inner_step": 260,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d020037a86842d26d2b202f08688e2577c805b23b1c18db1a5c9bbaea3e72e93
+oid sha256:537614b84509b3cdf9d6581fb29a584cd87696aca9a6030f95735105f6bed3ae
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb798c7be90ed7deb3c91241b732db42360ec1053cf35da0121e50d76d308cde
+oid sha256:0262d65772e080c39ac30618f793d6772131b883ab6bc1dbcc339ddd40e5d617
 size 4040701744